Pavan Kumar K

Optum – UnitedHealth Group

Data Engineering Consultant Jan 2017 – Present | Hyderabad, India

Consumer Experience & Relations (CXR) — ORS Curations & Train Stop Products (2022 – Present)

• Designed and deployed scalable data pipelines in Databricks and Snowflake, processing multi‑terabyte consumer interaction datasets daily.
• Curated Operational Reporting System (ORS) datasets, improving analytics accuracy for 1M+ members.
• Optimized train stop product ingestion and transformation workflows, reducing agent resolutions by calculating hops calages.
• Built reusable PySpark modules for aggregation and cleansing, reducing development cycles by 25%.

Optum Finance Advanced Intelligence Analytics (OFAIA) (2020 – 2022)

• Automated employer reporting workflows in Databricks, eliminating 90% of manual effort, and improving SLA compliance to 100%.
• Developed SQL and Python notebooks, processing 3+ TB of financial data per month, cutting report generation time from 6 hours to 45 minutes.
• Integrated Power BI dashboards that reduced stakeholder data request turnaround from days to same-day delivery.

Optum Finance Data & Analytics (OFDnA) (2018 – 2020)

• Created optimized views, stored procedures, and PySpark functions for enhanced data extraction and performance.
• Implemented static and dynamic Row Level Security, improving compliance and data governance.
• Reduced processing time by 30% through advanced data modeling and aggregation techniques.

Enterprise Decommissioning Program (EDP) (2017 – 2018)

• Led the migration of 20+ legacy healthcare platforms into the Azure Data Lakehouse, delivering $2M per year in cost savings.
• Built robust Spark ETL pipelines for 5+ TB/day data flows, with zero loss under strict compliance rules.
• Mentored a team of six junior engineers, accelerating project delivery by 40%.

Senior Data Engineer, 2014 – 2016. ||Hyderabad, India.

• Team: White Coats – a healthcare networking and engagement platform for doctors.
• Developed a robust medical search engine leveraging Elasticsearch and Logstash, improving search speed by 60% and accuracy by 25% for over 10,000 daily queries.
• Leveraged Neo4j Graph DB to design and implement a doctors’ networking graph, enabling intuitive relationship mapping, and increasing professional connection requests by 35%.
• Integrated graph-based queries with full-text search to enhance recommendations and user engagement.